CN116524725A - Intelligent driving traffic sign image data identification system - Google Patents
Intelligent driving traffic sign image data identification system Download PDFInfo
- Publication number
- CN116524725A CN116524725A CN202310803767.7A CN202310803767A CN116524725A CN 116524725 A CN116524725 A CN 116524725A CN 202310803767 A CN202310803767 A CN 202310803767A CN 116524725 A CN116524725 A CN 116524725A
- Authority
- CN
- China
- Prior art keywords
- traffic sign
- data
- image
- vehicle
- traffic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012545 processing Methods 0.000 claims abstract description 74
- 238000004458 analytical method Methods 0.000 claims abstract description 43
- 230000006855 networking Effects 0.000 claims abstract description 38
- 238000004891 communication Methods 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims description 30
- 239000011159 matrix material Substances 0.000 claims description 28
- 230000008569 process Effects 0.000 claims description 14
- 238000013500 data storage Methods 0.000 claims description 12
- 238000013499 data model Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/582—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of traffic signs
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an intelligent driving traffic sign image data identification system, which comprises an intelligent driving image processing layer and a third party application layer which establishes communication with the intelligent driving image processing layer through a wireless network, wherein the intelligent driving image processing layer comprises: the invention relates to the technical field of intelligent driving image processing, in particular to a central processing module, an image processing module, a vehicle real-time positioning module, a networking identification analysis unit and a historical traffic model creation unit. The intelligent driving traffic sign image data identification system can realize the more comprehensive and accurate identification analysis of the road traffic sign by organically combining the road traffic data acquired by the third party application with the road data acquired by the actual driving, and can perform normal identification under the condition that the traffic sign is partially absent or blocked by the back obstacle by combining the networking road and the traffic sign data, so that the partial incomplete traffic sign can be quickly and accurately identified.
Description
Technical Field
The invention relates to the technical field of intelligent driving image processing, in particular to a traffic sign image data identification system for intelligent driving.
Background
Road traffic sign recognition is taken as a basic branch of advanced driving assistance, is also an important means for improving traffic safety and efficiency, and mainly utilizes a vehicle-mounted special lens to be matched with a high-definition image photosensitive chip to collect traffic sign image data on a road. The data processor is adopted to perform cutting, denoising, standardization and other treatments on the acquired original image data so as to improve the accuracy and efficiency of subsequent identification. And extracting the characteristics of the image such as color, shape, texture and the like through a software algorithm, and extracting the characteristic information related to the traffic sign from the characteristics. And classifying the extracted characteristic information by using a machine learning algorithm, and distinguishing the extracted characteristic information into different traffic sign types. And according to the result of the feature classification, carrying out decision discrimination and output on the identified traffic sign so as to provide the traffic sign with an intelligent driving system to make a corresponding driving decision.
Referring to a traffic image recognition method, a device, computer equipment and a medium of Chinese patent publication No. CN109886210B, pictures in a video stream acquired by a vehicle are input to a self-interference elimination encoder, interference-filtered pictures are obtained through preprocessing of the self-interference elimination encoder, and then non-interference pictures are input to a traffic sign recognition model for recognition processing, so that a correct vehicle control instruction is conveniently generated subsequently, and the problem of traffic sign recognition errors caused by attack of a countersample on the traffic sign recognition model is solved; the interference of the countermeasures in the traffic image can be reduced, the identification accuracy of the image is improved, and the safety of unmanned driving or intelligent driving is improved. Referring to a method for constructing and identifying a traffic sign image identification model with a Chinese patent publication number of CN109002764B, a network model applied to traffic sign identification is obtained by training based on an improved ZF convolutional neural network and combining a space transformation network, so that the problem of misidentification caused by distortion and deformation of the traffic sign can be avoided, and the traffic sign identification rate is improved.
By comprehensively analyzing the above referenced patents, the following defects can be obtained: the existing traffic sign recognition system only relies on an image feature extraction, recognition and comparison analysis method to perform recognition and analysis on traffic signs in an acquired road view, and such recognition method cannot normally recognize the traffic signs in the condition of partial missing or partial blocking of a back obstacle, so that safe and intelligent driving of a vehicle cannot be ensured, for example, a traffic image recognition method, a device, computer equipment and medium of patent CN109886210B and a traffic sign image recognition model construction and recognition method of CN109002764B are referred to, and although clear recognition is performed on the traffic signs which are blurred in the acquired view by adopting a denoising and neural network feature training method, the traffic signs are missing in the aspect of partial missing, for example: when the traffic sign is used outdoors for a long time and is partially corroded to remove paint or is partially shielded by leaves beside a road, the traffic sign is processed only by virtue of the definition processing algorithm of the two references, the problem cannot be solved, the road traffic sign can not be comprehensively and accurately identified and analyzed by organically combining the road traffic data acquired by a third party application with the road data acquired by actual driving, the aim of establishing an identified complete traffic sign model by utilizing the networking road traffic sign data of the position where a vehicle is located to assist the data identified by actual driving can not be achieved, and therefore, the traffic sign with partial defect cannot be rapidly and accurately identified, and great inconvenience is brought to intelligent safe driving of people.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an intelligent driving traffic sign image data identification system, which solves the problems that the existing traffic sign identification system only relies on an image feature extraction, identification and comparison analysis method to identify and analyze traffic signs in an acquired road view, and the identification method cannot normally identify under the condition of partial missing of the traffic signs or partial shielding of back obstacles, so that the safe and intelligent driving of vehicles cannot be ensured.
In order to achieve the above purpose, the invention is realized by the following technical scheme: the utility model provides an intelligent driving's traffic sign image data identification system, includes intelligent driving image processing layer and establishes the third party's application layer of communication through wireless network with intelligent driving image processing layer, intelligent driving image processing layer includes:
the central processing module is used for controlling the whole image recognition system;
the view acquisition terminal is used for acquiring video image data of the surrounding environment of the automobile driving road in real time;
the image processing module is used for extracting, identifying and analyzing road traffic sign features in the video image acquired by the view acquisition terminal by adopting an image identification and analysis algorithm;
the vehicle real-time positioning module is used for positioning the position of the vehicle in real time;
the networking identification analysis unit is used for carrying out secondary identification contrast analysis processing on the image data analyzed by the image processing module through a networking auxiliary analysis algorithm, so that the image processing module carries out secondary analysis judgment and decision processing on the acquired partial incomplete traffic sign images;
the historical traffic model creation unit is used for creating a historical path traffic model on the road path traffic sign data processed by the view acquisition terminal, the image processing module, the vehicle implementation positioning module and the networking identification analysis unit in the intelligent running process of the vehicle through the data storage modeling algorithm, so that the historical path traffic model can be directly extracted for use in the historical path running process of the vehicle.
Preferably, the intelligent driving image processing layer further comprises a networking communication module, and the central processing module establishes wireless network communication with the third party application layer through the networking communication module, and is used for acquiring wireless road traffic data from the third party application layer.
Preferably, the intelligent driving image processing layer further comprises a data storage module, which is used for storing view data acquired by the view acquisition terminal in the intelligent driving process of the vehicle, data processed by the image processing module, vehicle driving track data positioned by the vehicle implementation positioning module, road path traffic sign data processed by the networking identification analysis unit and historical path traffic model data created by the historical traffic model creation unit.
Preferably, the view acquisition terminal is composed of n view acquisition modules, and the n view acquisition modules are arranged around the vehicle body and used for shooting and acquiring road traffic image data in the running process of the vehicle in real time.
Preferably, the networking auxiliary analysis algorithm specifically comprises the following steps:
s1, acquiring an actual road traffic sign characteristic data set obtained by recognition analysis of an image processing moduleWhere m is the number of images that are acquired and identified for analysis by the image processing moduleThe m-th traffic sign feature data matrix processed by the image processing module is specifically as follows:wherein: m represents the number of traffic sign types in the identified single image, n is the number of corresponding image texture feature values in each traffic sign of the single image,for a data set of m-th traffic sign data in a single image identified,the data of the texture characteristic value of the nth image in the mth traffic sign data for identifying the single image;
the acquired image processing module identifies and analyzes a data set consisting of traffic sign data in the single imageThe traffic sign image acquisition method comprises the steps of including a complete traffic sign feature data set and an incomplete traffic sign feature data set, wherein the incomplete traffic sign feature data set is acquired and identified partial incomplete traffic sign image features;
s2, acquiring the position of the vehicle positioned by the vehicle real-time positioning moduleThe data information and the networking traffic sign characteristic data set corresponding to the vehicle position data information are acquired from a third party application layerWhere d is the number of networked images corresponding to the locationThe feature data matrix of the traffic sign for the d-th traffic sign acquired from the third party application layer is specifically as follows:wherein: d represents the number of traffic sign types in the acquired single image of the vehicle position, e is the number of corresponding image texture feature values in each traffic sign of the single image of the vehicle position acquired from the third party application layer,to obtain a data set of d-th traffic sign data in a single image of this vehicle location from a third party application layer,the method comprises the steps of acquiring the e-th image texture characteristic value data in the d-th traffic sign data of a single image of the vehicle position from a third party application layer;
s3, identifying and analyzing the traffic sign characteristic data set obtained by the image processing module obtained in the step S1Each piece of complete traffic sign feature data matrix feature data in the network traffic sign feature data set corresponding to the vehicle position data information is acquired from a third party application layer at the moment in the step S2Performing difference comparison on each piece of traffic sign feature data matrix feature data, and then classifying the comparison result according to two types of complete matching and complete non-matching;
s4, when the pair is toData sets consisting of traffic sign data with perfect match of comparison results, i.e. When the comparison result shows that the traffic sign in the acquired image is matched with the traffic sign of the position of the network acquired vehicle, the vehicle executes corresponding driving instructions according to the traffic sign corresponding to the traffic sign data set consisting of the completely matched traffic sign data, and when the comparison result has the completely unmatched traffic sign data set, namelyAndwhen the traffic signs are completely unequal, the traffic signs in the acquired images are not matched with the traffic signs of the vehicle positions acquired through networking, and the vehicle executes corresponding intelligent driving instructions according to the traffic signs of the actual road traffic sign feature data set acquired through the identification and analysis of the acquired image processing module in the step S1; s5, identifying and analyzing the traffic sign characteristic data set obtained by the image processing module obtained in the step S1Each piece of incomplete traffic sign feature data matrix feature data in the network traffic sign feature data set corresponding to the vehicle position data information is acquired from a third party application layer at the moment in the step S2Each traffic sign characteristic data matrix characteristic data in the image is subjected to difference comparison, so that the acquired incompletely identified traffic sign data set in a single imageIn (a) and (b)Individual feature data and traffic sign data set in a single image of the vehicle location obtained from a third party application layerThe difference is calculated among the characteristic data of the vehicle, and then the traffic sign data set in each single image for acquiring the vehicle position from the third party application layer is countedThe number a of the same characteristic value in the vehicle is the traffic sign data set in a single image obtained from a third party application layerThe percentage of the total number b of the characteristic values is the similarity delta, and the specific formula is as follows:s6, sorting the similarity delta obtained in the step S5, and extracting a group of traffic sign data sets in a single image of the vehicle position obtained from a third party application layer with the highest similarity from the sequence as a preselected feature setThen judge the preselect feature setWhether the pre-selected feature set is larger than the critical similarity beta, if so, determining the pre-selected feature setFor target feature set, incomplete identified traffic sign data set in single imageThe characteristic values in the target characteristic set are replaced, so that the collected partial incomplete traffic sign images are completed, and then the identification operation of the complete matching type in the step S3 is carried out;
s7, if the preset characteristic set is smaller than beta, judging the preset characteristic setAnd if the traffic sign is the non-target feature set, namely, judging that the accurate traffic sign is not recognized, and then executing a corresponding intelligent driving instruction by the vehicle according to the traffic sign of the networking traffic sign feature data set corresponding to the vehicle position data information acquired from the third party application layer at the moment.
Preferably, in the step S3, the traffic sign feature data matrix of each image is compared with the traffic sign feature data matrix of the time and the position acquired from the third party application layer, that is, the judgment is performedAndwhether or not the feature data are identical, where m=d.
Preferably, in the step S3, the traffic sign feature data matrix in each image is compared with each otherAnd (3) comparing the characteristic data in the data set formed by each traffic sign data one by one, and judging whether the characteristic data are identical.
Preferably, the value range of the critical similarity β% in the step S6 is 30% -60%.
Preferably, the data storage modeling algorithm is one of a hash algorithm or a tree structure algorithm.
The invention provides an intelligent driving traffic sign image data identification system. Compared with the prior art, the method has the following beneficial effects:
(1) The intelligent driving traffic sign image data identification system can realize the purpose of carrying out more comprehensive and accurate identification and analysis on the road traffic sign by organically combining the road traffic data acquired by a third party application with the road data acquired by actual driving, well achieves the purpose of establishing an identified complete traffic sign model by utilizing the networking road traffic sign data of the position where the vehicle is located to assist the data identified by actual driving, and carries out normal identification on the condition that part of the traffic sign is missing or is blocked by a back obstacle by combining networking road and traffic sign data on the basis of image feature extraction, identification and comparison analysis, thereby ensuring safe and intelligent driving of the vehicle, and carrying out intelligent judgment and analysis processing even if the situation that the traffic sign is corroded by part or blocked by leaves beside the road for a long time is met, thereby rapidly and accurately identifying the part of the traffic sign, and greatly facilitating intelligent and safe driving of people.
(2) The intelligent driving traffic sign image data identification system can realize the creation of a route sign model through data information which is driven by a user and acquired networking road information corresponding to the data information, and store the data information into a database, so that when a later user drives along the route again, the corresponding route traffic sign model is only required to be extracted from the database to compare and analyze with the actually acquired road sign information, the purpose of quickly and accurately acquiring the traffic sign information data of the passing route without repetition is well achieved, the condition that the acquisition time is long or the acquisition fails due to the influence of traffic road environment factors when the road traffic sign information is acquired by networking is avoided, the condition that the on-site traffic sign identification is assisted by acquiring the road traffic sign information by a third party through networking is not normally realized, and the history data in a networking-free state is realized through directly calling the history track of the vehicle and the traffic sign data information corresponding to the history track, so that the intelligent driving is very beneficial to the intelligent driving in a network environment.
Drawings
Fig. 1 is a schematic block diagram of the system of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, two technical schemes are provided in the embodiment of the present invention: an intelligent driving traffic sign image data identification system specifically comprises the following embodiments:
the utility model provides an intelligent driving's traffic sign image data identification system, includes intelligent driving image processing layer and establishes the third party's application layer of communication through wireless network with intelligent driving image processing layer, intelligent driving image processing layer includes:
the central processing module is used for controlling the whole image recognition system;
the view acquisition terminal is used for acquiring video image data of the surrounding environment of the automobile driving road in real time;
the image processing module is used for extracting, identifying and analyzing road traffic sign features in the video image acquired by the view acquisition terminal by adopting an image identification and analysis algorithm;
the vehicle real-time positioning module is used for positioning the position of the vehicle in real time;
the networking identification analysis unit is used for carrying out secondary identification contrast analysis processing on the image data analyzed by the image processing module through a networking auxiliary analysis algorithm, so that the image processing module carries out secondary analysis judgment and decision processing on the acquired partial incomplete traffic sign images;
the historical traffic model creation unit is used for creating a historical path traffic model on the road path traffic sign data processed by the view acquisition terminal, the image processing module, the vehicle implementation positioning module and the networking identification analysis unit in the intelligent running process of the vehicle through the data storage modeling algorithm, so that the historical path traffic model can be directly extracted for use in the historical path running process of the vehicle.
In the embodiment of the invention, the intelligent driving image processing layer further comprises a networking communication module, the central processing module establishes wireless network communication with the third party application layer through the networking communication module, and is used for acquiring wireless road traffic data from the third party application layer, and the intelligent driving image processing layer further comprises a data storage module for storing view data acquired by the view acquisition terminal, data processed by the image processing module, vehicle driving track data positioned by the vehicle implementation positioning module, road path traffic sign data processed by the networking identification analysis unit and historical path traffic model data created by the historical traffic model creation unit in the intelligent driving process of the vehicle.
In the embodiment of the invention, the view acquisition terminal consists of n view acquisition modules, and the n view acquisition modules are arranged around the vehicle body and are used for shooting and acquiring road traffic image data in the running process of the vehicle in real time.
In the embodiment of the invention, the networking auxiliary analysis algorithm specifically comprises the following steps:
s1, acquiring an actual road traffic sign characteristic data set obtained by recognition analysis of an image processing moduleWhere m is the number of images that are acquired and identified for analysis by the image processing moduleThe m-th traffic sign feature data matrix processed by the image processing module is specifically as follows:wherein: m represents the number of traffic sign types in the identified single image, n is the number of corresponding image texture feature values in each traffic sign of the single image,for a data set of m-th traffic sign data in a single image identified,nth graph in mth traffic sign data for identifying single imageTexture feature value data;
the acquired image processing module identifies and analyzes a data set consisting of traffic sign data in the single imageThe traffic sign image acquisition method comprises the steps of including a complete traffic sign feature data set and an incomplete traffic sign feature data set, wherein the incomplete traffic sign feature data set is acquired and identified partial incomplete traffic sign image features;
s2, acquiring the position data information of the vehicle positioned by the vehicle real-time positioning module, and acquiring a networking traffic sign characteristic data set corresponding to the position data information of the vehicle from a third party application layerWhere d is the number of networked images corresponding to the locationThe feature data matrix of the traffic sign for the d-th traffic sign acquired from the third party application layer is specifically as follows:wherein: d represents the number of traffic sign types in the acquired single image of the vehicle position, e is the number of corresponding image texture feature values in each traffic sign of the single image of the vehicle position acquired from the third party application layer,to obtain a data set of d-th traffic sign data in a single image of this vehicle location from a third party application layer,the method comprises the steps of acquiring the e-th image texture characteristic value data in the d-th traffic sign data of a single image of the vehicle position from a third party application layer;
s3, identifying and analyzing the traffic sign characteristic data set obtained by the image processing module obtained in the step S1Each piece of complete traffic sign feature data matrix feature data in the network traffic sign feature data set corresponding to the vehicle position data information is acquired from a third party application layer at the moment in the step S2Performing difference comparison on each piece of traffic sign feature data matrix feature data, and then classifying the comparison result according to two types of complete matching and complete non-matching;
s4, when the comparison result has a data set composed of traffic sign data which are completely matched, namely When the comparison result shows that the traffic sign in the acquired image is matched with the traffic sign of the position of the network acquired vehicle, the vehicle executes corresponding driving instructions according to the traffic sign corresponding to the traffic sign data set consisting of the completely matched traffic sign data, and when the comparison result has the completely unmatched traffic sign data set, namelyAndwhen the traffic signs are completely unequal, the traffic signs in the acquired images are not matched with the traffic signs of the vehicle positions acquired through networking, and the vehicle executes corresponding intelligent driving instructions according to the traffic signs of the actual road traffic sign feature data set acquired through the identification and analysis of the acquired image processing module in the step S1; s5, identifying and analyzing the traffic sign characteristic data set obtained by the image processing module obtained in the step S1Each piece of incomplete traffic sign feature data matrix feature data in the network traffic sign feature data set corresponding to the vehicle position data information is acquired from a third party application layer at the moment in the step S2Each traffic sign characteristic data matrix characteristic data in the image is subjected to difference comparison, so that the acquired incompletely identified traffic sign data set in a single imageIn a single image of the vehicle location obtained from a third party application layerThe difference is calculated among the characteristic data of the vehicle, and then the traffic sign data set in each single image for acquiring the vehicle position from the third party application layer is countedThe number a of the same characteristic value in the vehicle is the traffic sign data set in a single image obtained from a third party application layerThe percentage of the total number b of the characteristic values is the similarity delta, and the specific formula is as follows:s6, sorting the similarity delta obtained in the step S5, and extracting a group of traffic sign data sets in a single image of the vehicle position obtained from a third party application layer with the highest similarity from the sequence as a preselected feature setThen judge the preselect feature setWhether the pre-selected feature set is larger than the critical similarity beta, if so, determining the pre-selected feature setFor target feature set, incomplete identified traffic sign data set in single imageThe characteristic values in the target characteristic set are replaced, so that the collected partial incomplete traffic sign images are completed, and then the identification operation of the complete matching type in the step S3 is carried out;
s7, if the preset characteristic set is smaller than beta, judging the preset characteristic setAnd if the traffic sign is the non-target feature set, namely, judging that the accurate traffic sign is not recognized, and then executing a corresponding intelligent driving instruction by the vehicle according to the traffic sign of the networking traffic sign feature data set corresponding to the vehicle position data information acquired from the third party application layer at the moment.
In the embodiment of the invention, in step S3, the traffic sign characteristic data matrix of each image is compared with the traffic sign characteristic data matrix of the moment and the position acquired from the third party application layer, namely judgmentAndwhether or not the feature data are identical, where m=d.
In the embodiment of the invention, in the step S3, when the traffic sign characteristic data matrix in each image is compared, the traffic sign characteristic data matrix is displayedAnd (3) comparing the characteristic data in the data set formed by each traffic sign data one by one, and judging whether the characteristic data are identical.
In the embodiment of the invention, in the step S6, the value range of the critical similarity beta% is 30% -60%, and the specific acquaintance value is set according to the actual use requirement of the user.
In the embodiment of the invention, the image recognition analysis algorithm is an existing OCR algorithm, and specifically comprises the following steps:
1. and (3) inputting an image: there are different storage formats and different compression modes for different image formats.
Binarization: most of pictures shot by a camera are color images, the information content of the color images is huge, the content of the pictures can be simply divided into foreground and background, in order to enable a computer to quickly and better identify the characteristics in the images, the color images need to be processed first, only foreground information and background information of the pictures are enabled to be simply defined, the foreground information can be black, the background information can be white, and the binarization images can be generated.
Noise removal: the definition of the noise can be different for different images, and the noise is removed according to the characteristics of the noise, which is called noise removal, and specifically includes defogging and definition processing.
Tilt alignment: because the vehicle runs on a bumpy road or the traffic sign itself is inclined under external factors, the collected pictures are inevitably inclined, and the pictures are required to be corrected through an identification algorithm.
Graph analysis: the collected picture is divided into a plurality of parts according to the characteristic points so as to extract the traffic sign graphic characteristics.
Identification contrast: the identified plurality of feature data are extracted from the graph and compared with data in a database pre-stored with all road traffic signs.
Post-processing and checking: and (3) according to a specific recognition environment (rainy days, foggy days or night), correcting the recognition result.
In the embodiment of the invention, the data storage modeling algorithm is an existing hash algorithm.
The hash algorithm is an algorithm for mapping data into a table with a fixed size, the data is converted into a unique hash value through a hash function, then the hash value and a corresponding table position are mapped, so that quick data searching and accessing are realized, the traffic sign data processed by the system are imported into the data, a plurality of historical traffic data models are created in a data storage module through a historical traffic model creation unit, and each stored data model can be quickly searched and accessed through a data query algorithm of the hash algorithm.
Example 2
Compared with the embodiment 1, the distinguishing technical scheme of the embodiment is as follows: the data storage modeling algorithm is an existing tree structure algorithm.
The tree structure algorithm organizes data by using a tree structure, sorts the data according to a certain rule, and then sequentially inserts the sorted data into a tree. And then the traffic sign data processed by the system is imported into the data, a plurality of historical traffic data models are created in the data storage module through the historical traffic model creation unit, each stored data model can be quickly searched and accessed through a data query algorithm of a hash algorithm, and compared with the existing hash algorithm of the embodiment, the existing tree structure algorithm adopted by the embodiment has the advantages that the search and insertion operation can be quickly performed, and better performance can be kept when the data quantity is larger.
And all that is not described in detail in this specification is well known to those skilled in the art.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. The utility model provides a traffic sign image data identification system of intelligent driving, includes intelligent driving image processing layer and establishes the third party application layer of communication through wireless network with intelligent driving image processing layer, its characterized in that: the intelligent driving image processing layer comprises:
the central processing module is used for controlling the whole image recognition system;
the view acquisition terminal is used for acquiring video image data of the surrounding environment of the automobile driving road in real time;
the image processing module is used for extracting, identifying and analyzing road traffic sign features in the video image acquired by the view acquisition terminal by adopting an image identification and analysis algorithm;
the vehicle real-time positioning module is used for positioning the position of the vehicle in real time;
the networking identification analysis unit is used for carrying out secondary identification contrast analysis processing on the image data analyzed by the image processing module through a networking auxiliary analysis algorithm, so that the image processing module carries out secondary analysis judgment and decision processing on the acquired partial incomplete traffic sign images;
the historical traffic model creation unit is used for creating a historical path traffic model on the road path traffic sign data processed by the view acquisition terminal, the image processing module, the vehicle implementation positioning module and the networking identification analysis unit in the intelligent running process of the vehicle through the data storage modeling algorithm, so that the historical path traffic model can be directly extracted for use in the historical path running process of the vehicle.
2. The intelligent driving traffic sign image data recognition system according to claim 1, wherein: the intelligent driving image processing layer further comprises a networking communication module, and the central processing module establishes wireless network communication with the third party application layer through the networking communication module and is used for acquiring wireless road traffic data from the third party application layer.
3. The intelligent driving traffic sign image data recognition system according to claim 1, wherein: the intelligent driving image processing layer also comprises a data storage module which is used for storing view data acquired by the view acquisition terminal in the intelligent driving process of the vehicle, data processed by the image processing module, vehicle driving track data positioned by the vehicle implementation positioning module, road path traffic sign data processed by the networking identification analysis unit and historical path traffic model data created by the historical traffic model creation unit.
4. The intelligent driving traffic sign image data recognition system according to claim 1, wherein: the view acquisition terminal consists of n view acquisition modules, and the n view acquisition modules are arranged around the vehicle body and are used for shooting and acquiring road traffic image data in the running process of the vehicle in real time.
5. The intelligent driving traffic sign image data recognition system according to claim 1, wherein: the networking auxiliary analysis algorithm specifically comprises the following steps:
s1, acquiring an actual road traffic sign characteristic data set obtained by recognition analysis of an image processing moduleWhere m is the number of images that are acquired and identified for analysis by the image processing moduleThe m-th traffic sign feature data matrix processed by the image processing module is specifically as follows:
wherein: m represents the number of traffic sign types in the identified single image, n is the number of corresponding image texture feature values in each traffic sign of the single image,for a data set of m-th traffic sign data in a single image identified,the data of the texture characteristic value of the nth image in the mth traffic sign data for identifying the single image;
the acquired image processing module identifies and analyzes a data set consisting of traffic sign data in the single imageThe traffic sign image acquisition method comprises the steps of including a complete traffic sign feature data set and an incomplete traffic sign feature data set, wherein the incomplete traffic sign feature data set is acquired and identified partial incomplete traffic sign image features;
s2, acquiring the position data information of the vehicle positioned by the vehicle real-time positioning module, and acquiring a networking traffic sign characteristic data set corresponding to the position data information of the vehicle from a third party application layerWhere d is the number of networked images corresponding to the locationThe feature data matrix of the traffic sign for the d-th traffic sign acquired from the third party application layer is specifically as follows:wherein: d represents the number of traffic sign types in the acquired single image of the vehicle position, e is the texture of the corresponding image in each traffic sign of the single image of the vehicle position acquired from the third party application layerThe number of the sign values,to obtain a data set of d-th traffic sign data in a single image of this vehicle location from a third party application layer,the method comprises the steps of acquiring the e-th image texture characteristic value data in the d-th traffic sign data of a single image of the vehicle position from a third party application layer;
s3, identifying and analyzing the traffic sign characteristic data set obtained by the image processing module obtained in the step S1Each piece of complete traffic sign feature data matrix feature data in the network traffic sign feature data set corresponding to the vehicle position data information is acquired from a third party application layer at the moment in the step S2Performing difference comparison on each piece of traffic sign feature data matrix feature data, and then classifying the comparison result according to two types of complete matching and complete non-matching;
s4, when the comparison result has a data set composed of traffic sign data which are completely matched, namely When the comparison result shows that the traffic sign in the acquired image is matched with the traffic sign of the position of the network acquired vehicle, the vehicle executes corresponding driving instructions according to the traffic sign corresponding to the data set consisting of the completely matched traffic sign data, and when the comparison result has the completely unmatched traffic sign numberFrom a composed data set, i.e.Andwhen the traffic signs are completely unequal, the traffic signs in the acquired images are not matched with the traffic signs of the vehicle positions acquired through networking, and the vehicle executes corresponding intelligent driving instructions according to the traffic signs of the actual road traffic sign feature data set acquired through the identification and analysis of the acquired image processing module in the step S1;
s5, identifying and analyzing the traffic sign characteristic data set obtained by the image processing module obtained in the step S1Each piece of incomplete traffic sign feature data matrix feature data in the network traffic sign feature data set corresponding to the vehicle position data information is acquired from a third party application layer at the moment in the step S2Each traffic sign characteristic data matrix characteristic data in the image is subjected to difference comparison, so that the acquired incompletely identified traffic sign data set in a single imageIn a single image of the vehicle location obtained from a third party application layerThe difference is calculated among the characteristic data of the vehicle, and then the traffic sign data set in each single image for acquiring the vehicle position from the third party application layer is countedThe number a of the same characteristic value in the vehicle is the traffic sign in a single image obtained from a third party application layerSets of lineage dataThe percentage of the total number b of the characteristic values is the similarity delta, and the specific formula is as follows:
s6, sorting the similarity delta obtained in the step S5, and extracting a group of traffic sign data sets in a single image of the vehicle position obtained from a third party application layer with the highest similarity from the sequence as a preselected feature setThen judge the preselect feature setWhether the pre-selected feature set is larger than the critical similarity beta, if so, determining the pre-selected feature setFor target feature set, incomplete identified traffic sign data set in single imageThe characteristic values in the target characteristic set are replaced, so that the collected partial incomplete traffic sign images are completed, and then the identification operation of the complete matching type in the step S3 is carried out;
s7, if the preset characteristic set is smaller than beta, judging the preset characteristic setAnd if the traffic sign is the non-target feature set, namely, judging that the accurate traffic sign is not recognized, and then executing a corresponding intelligent driving instruction by the vehicle according to the traffic sign of the networking traffic sign feature data set corresponding to the vehicle position data information acquired from the third party application layer at the moment.
6. The intelligent driving traffic sign image data recognition system according to claim 5, wherein: the traffic sign characteristic data matrix of each image is compared with the traffic sign characteristic data matrix of the moment and the position acquired from the third party application layer in the step S3, namely judgmentAndwhether or not the feature data are identical, where m=d.
7. The intelligent driving traffic sign image data recognition system according to claim 6, wherein: in the step S3, when the traffic sign characteristic data matrix in each image is comparedAnd (3) comparing the characteristic data in the data set formed by each traffic sign data one by one, and judging whether the characteristic data are identical.
8. The intelligent driving traffic sign image data recognition system according to claim 1, wherein: and in the step S6, the value range of the critical similarity beta% is 30% -60%.
9. The intelligent driving traffic sign image data recognition system according to claim 1, wherein: the data storage modeling algorithm is one of a hash algorithm or a tree structure algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310803767.7A CN116524725B (en) | 2023-07-03 | 2023-07-03 | Intelligent driving traffic sign image data identification system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310803767.7A CN116524725B (en) | 2023-07-03 | 2023-07-03 | Intelligent driving traffic sign image data identification system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116524725A true CN116524725A (en) | 2023-08-01 |
CN116524725B CN116524725B (en) | 2023-09-01 |
Family
ID=87406704
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310803767.7A Active CN116524725B (en) | 2023-07-03 | 2023-07-03 | Intelligent driving traffic sign image data identification system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116524725B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116778457A (en) * | 2023-08-16 | 2023-09-19 | 钧捷科技(北京)有限公司 | Automatic parking auxiliary control system and device for vehicle |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102819728A (en) * | 2012-07-17 | 2012-12-12 | 中国航天科工集团第三研究院第八三五七研究所 | Traffic sign detection method based on classification template matching |
CN104008377A (en) * | 2014-06-07 | 2014-08-27 | 北京联合大学 | Ground traffic sign real-time detection and recognition method based on space-time correlation |
US20200324787A1 (en) * | 2018-10-25 | 2020-10-15 | Samsung Electronics Co., Ltd. | Augmented reality method and apparatus for driving assistance |
CN114821519A (en) * | 2022-03-21 | 2022-07-29 | 上海应用技术大学 | Traffic sign identification method and system based on coordinate attention |
CN116259040A (en) * | 2023-03-20 | 2023-06-13 | 智道网联科技(北京)有限公司 | Method and device for identifying traffic sign and electronic equipment |
-
2023
- 2023-07-03 CN CN202310803767.7A patent/CN116524725B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102819728A (en) * | 2012-07-17 | 2012-12-12 | 中国航天科工集团第三研究院第八三五七研究所 | Traffic sign detection method based on classification template matching |
CN104008377A (en) * | 2014-06-07 | 2014-08-27 | 北京联合大学 | Ground traffic sign real-time detection and recognition method based on space-time correlation |
US20200324787A1 (en) * | 2018-10-25 | 2020-10-15 | Samsung Electronics Co., Ltd. | Augmented reality method and apparatus for driving assistance |
CN114821519A (en) * | 2022-03-21 | 2022-07-29 | 上海应用技术大学 | Traffic sign identification method and system based on coordinate attention |
CN116259040A (en) * | 2023-03-20 | 2023-06-13 | 智道网联科技(北京)有限公司 | Method and device for identifying traffic sign and electronic equipment |
Non-Patent Citations (1)
Title |
---|
唐智威 等: "基于视觉的无人驾驶汽车研究综述", 制造业自动化, no. 08 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116778457A (en) * | 2023-08-16 | 2023-09-19 | 钧捷科技(北京)有限公司 | Automatic parking auxiliary control system and device for vehicle |
CN116778457B (en) * | 2023-08-16 | 2023-11-03 | 钧捷科技(北京)有限公司 | Automatic parking auxiliary control system and device for vehicle |
Also Published As
Publication number | Publication date |
---|---|
CN116524725B (en) | 2023-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112686812B (en) | Bank card inclination correction detection method and device, readable storage medium and terminal | |
CN110728225B (en) | High-speed face searching method for attendance checking | |
CN106610969A (en) | Multimodal information-based video content auditing system and method | |
CN112381075B (en) | Method and system for carrying out face recognition under specific scene of machine room | |
CN110008909B (en) | Real-name system business real-time auditing system based on AI | |
CN111488875A (en) | Vehicle insurance claim settlement loss checking method and device based on image recognition and electronic equipment | |
CN109697440A (en) | A kind of ID card information extracting method | |
CN111310662B (en) | Flame detection and identification method and system based on integrated deep network | |
CN111382690B (en) | Vehicle re-identification method based on multi-loss fusion model | |
CN116524725B (en) | Intelligent driving traffic sign image data identification system | |
CN108323209A (en) | Information processing method, system, cloud processing device and computer program product | |
CN110738100A (en) | camouflage military target identification method and system based on deep learning | |
CN112784834A (en) | Automatic license plate identification method in natural scene | |
CN110852327A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
CN113971821A (en) | Driver information determination method and device, terminal device and storage medium | |
CN114067286A (en) | High-order camera vehicle weight recognition method based on serialized deformable attention mechanism | |
CN112784754A (en) | Vehicle re-identification method, device, equipment and storage medium | |
CN115761888A (en) | Tower crane operator abnormal behavior detection method based on NL-C3D model | |
CN116664944A (en) | Vineyard pest identification method based on attribute feature knowledge graph | |
CN110321890B (en) | Digital instrument identification method of power inspection robot | |
CN111178405A (en) | Similar object identification method fusing multiple neural networks | |
CN109255052B (en) | Three-stage vehicle retrieval method based on multiple features | |
CN117292338B (en) | Vehicle accident identification and analysis method based on video stream analysis | |
CN111832497B (en) | Text detection post-processing method based on geometric features | |
CN112966730A (en) | Vehicle damage identification method, device, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |